MiniCPM5-2B
73 records across 10 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M5 Pro | MiniCPM5-2B | oQ4e | oMLX | 136.3 | 3830.0 | – | – | 4096 | 2026-10-17 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 132.4 | 1565.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 132.4 | 1565.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 132.4 | 1565.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 132.4 | 1565.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 116.2 | 1423.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 116.2 | 1423.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 116.2 | 1423.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 116.2 | 1423.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 105.2 | 1181.0 | – | – | 8192 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 105.2 | 1181.0 | – | – | 8192 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 105.2 | 1181.0 | – | – | 8192 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 105.2 | 1181.0 | – | – | 8192 | 2026-10-06 | community | – | source |
| M5 Pro | MiniCPM5-2B | oQ4e | oMLX | 102.5 | 3877.0 | – | – | 16384 | 2026-10-17 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 101.6 | 2858.0 | – | – | 8192 | 2026-10-10 | community | – | source |
| M1 Max | MiniCPM5-2B | 4bit | oMLX | 93.4 | 866.8 | – | – | 4096 | 2026-10-13 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 90.6 | 2671.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 89.3 | 886.3 | – | – | 16384 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 89.3 | 886.3 | – | – | 16384 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 89.3 | 886.3 | – | – | 16384 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 89.3 | 886.3 | – | – | 16384 | 2026-10-06 | community | – | source |
| M1 Max | MiniCPM5-2B | oQ4e | oMLX | 88.4 | 876.9 | – | – | 4096 | 2026-10-13 | community | – | source |
| M1 Max | MiniCPM5-2B | oQ4e | oMLX | 88.3 | 890.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M5 Max | MiniCPM5-2B | 4bit | oMLX | 87.5 | 2901.0 | – | – | 65536 | 2026-10-06 | community | – | source |
| M5 Max | MiniCPM5-2B | 4bit | oMLX | 87.5 | 2901.0 | – | – | 65536 | 2026-10-06 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 83.8 | 2857.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 80.4 | 3192.0 | – | – | 65536 | 2026-10-10 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 78.7 | 2753.0 | – | – | 16384 | 2026-10-10 | community | – | source |
| M1 Pro | MiniCPM5-2B | 4bit | oMLX | 74.2 | 667.6 | – | – | 1024 | 2026-10-14 | community | – | source |
| M1 Pro | MiniCPM5-2B | 4bit | oMLX | 74.2 | 654.9 | – | – | 1024 | 2026-10-15 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 70.3 | 1848.0 | – | – | 1024 | 2026-10-07 | community | – | source |
| M1 Max | MiniCPM5-2B | 4bit | oMLX | 66.2 | 616.0 | – | – | 16384 | 2026-10-13 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 65.8 | 621.7 | – | – | 32768 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 65.8 | 621.7 | – | – | 32768 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 65.8 | 621.7 | – | – | 32768 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | oQ4e | oMLX | 65.8 | 621.7 | – | – | 32768 | 2026-10-06 | community | – | source |
| M1 Pro | MiniCPM5-2B | 4bit | oMLX | 64.3 | 610.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 Pro | MiniCPM5-2B | 4bit | oMLX | 64.0 | 610.0 | – | – | 4096 | 2026-10-14 | community | – | source |
| M5 Max | MiniCPM5-2B | 8bit | oMLX | 63.8 | 2367.0 | – | – | 32768 | 2026-10-10 | community | – | source |
| M1 Max | MiniCPM5-2B | oQ4e | oMLX | 63.7 | 615.3 | – | – | 16384 | 2026-10-13 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 63.6 | 1854.0 | – | – | 4096 | 2026-10-07 | community | – | source |
| M1 Max | MiniCPM5-2B | oQ4e | oMLX | 63.6 | 615.3 | – | – | 16384 | 2026-10-13 | community | – | source |
| M3 Ultra | MiniCPM5-2B | unknown | oMLX | 63.0 | 1076.0 | – | – | 65536 | 2026-10-08 | community | – | source |
| M3 Ultra | MiniCPM5-2B | bf16 | oMLX | 62.5 | 1076.0 | – | – | 65536 | 2026-10-08 | community | – | source |
| M4 Pro | MiniCPM5-2B | unknown | oMLX | 50.2 | 1619.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | unknown | oMLX | 50.2 | 1619.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | unknown | oMLX | 47.8 | 1526.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | MiniCPM5-2B | unknown | oMLX | 47.8 | 1526.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 | MiniCPM5-2B | 8bit | oMLX | 39.0 | 756.9 | – | – | 1024 | 2026-10-17 | community | – | source |
| M1 | MiniCPM5-2B | oQ4e | oMLX | 37.3 | 326.5 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 Max (128 GB) | MiniCPM5-2B | Q8_0 | llama.cpp | 35.5 | – | – | – | 4096 | 2026-10-17 | sourced | – | source |
| M4 | MiniCPM5-2B | 8bit | oMLX | 35.5 | 700.5 | – | – | 4096 | 2026-10-17 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 35.3 | 1100.0 | – | – | 16384 | 2026-10-14 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 33.8 | 1220.0 | – | – | 8192 | 2026-10-13 | community | – | source |
| M1 | MiniCPM5-2B | 4bit | oMLX | 33.6 | 313.5 | – | – | 1024 | 2026-10-10 | community | – | source |
| M5 Max | MiniCPM5-2B | 4bit | oMLX | 32.5 | 991.2 | – | – | 199680 | 2026-10-06 | community | – | source |
| M5 Max | MiniCPM5-2B | 4bit | oMLX | 32.5 | 991.2 | – | – | 199680 | 2026-10-06 | community | – | source |
| M1 | MiniCPM5-2B | oQ4e | oMLX | 32.1 | 290.6 | – | – | 4096 | 2026-10-08 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 31.5 | 951.2 | – | – | 16384 | 2026-10-13 | community | contradiction | source |
| M1 | MiniCPM5-2B | 4bit | oMLX | 26.9 | 292.5 | – | – | 4096 | 2026-10-10 | community | – | source |
| M1 | MiniCPM5-2B | oQ5 | oMLX | 26.6 | 404.6 | – | – | 1024 | 2026-10-07 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 26.3 | 827.8 | – | – | 32768 | 2026-10-14 | community | – | source |
| M5 | MiniCPM5-2B | 4bit | oMLX | 22.0 | 707.6 | – | – | 32768 | 2026-10-13 | community | contradiction | source |
| M4 | MiniCPM5-2B | bf16 | oMLX | 21.2 | 807.4 | – | – | 1024 | 2026-10-15 | community | – | source |
| M4 | MiniCPM5-2B | bf16 | oMLX | 21.1 | 797.1 | – | – | 1024 | 2026-10-14 | community | – | source |
| M4 Max (128 GB) | MiniCPM5-2B | Q8_0 | llama.cpp | 20.6 | – | – | – | 32768 | 2026-10-17 | sourced | – | source |
| M4 | MiniCPM5-2B | bf16 | oMLX | 20.1 | 760.9 | – | – | 4096 | 2026-10-14 | community | – | source |
| M4 | MiniCPM5-2B | bf16 | oMLX | 20.0 | 763.7 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 | MiniCPM5-2B | 4bit | oMLX | 19.4 | 247.1 | – | – | 8192 | 2026-10-10 | community | – | source |
| M1 | MiniCPM5-2B | oQ5 | oMLX | 16.5 | 333.1 | – | – | 8192 | 2026-10-07 | community | – | source |
| M1 | MiniCPM5-2B | 4bit | oMLX | 15.5 | 190.8 | – | – | 16384 | 2026-10-10 | community | – | source |
| M4 Max (128 GB) | MiniCPM5-2B | Q8_0 | llama.cpp | 11.4 | – | – | – | 102400 | 2026-10-17 | sourced | – | source |
| M1 | MiniCPM5-2B | 4bit | oMLX | 10.1 | 133.4 | – | – | 32768 | 2026-10-10 | community | – | source |